Figure - available from: Journal of Sensors
This content is subject to copyright. Terms and conditions apply.
Source publication
Biometric face recognition is becoming more frequently used in different application scenarios. However, spoofing attacks with facial disguises are still a serious problem for state of the art face recognition algorithms. This work proposes an approach to face verification based on spectral signatures of material surfaces in the short wave infrared...
Similar publications
Recent years have seen growing efforts to develop spoofing countermeasures (CMs) to protect automatic speaker verification (ASV) systems from being deceived by manipulated or artificial inputs. The reliability of spoofing CMs is typically gauged using the equal error rate (EER) metric. The primitive EER fails to reflect application requirements and...
Recent years have seen growing efforts to develop spoofing countermeasures (CMs) to protect automatic speaker verification (ASV) systems from being deceived by manipulated or artificial inputs. The reliability of spoofing CMs is typically gauged using the equal error rate (EER) metric. The primitive EER fails to reflect application requirements and...
![Fig. 1 Sample images from the PRINT-ATTACK [1] and REPLAY-ATTACK [12]...](profile/Jukka-Komulainen/publication/330400888/figure/fig1/AS:715433976020995@1547584160983/Sample-images-from-the-PRINT-ATTACK-1-and-REPLAY-ATTACK-12-databases-Top-and-bottom_Q320.jpg)
Face presentation attack detection has received increasing attention ever since the vulnerabilities to spoofing have been widely recognized. The state of the art in software-based face anti-spoofing has been assessed in three international competitions organized in conjunction with major biometrics conferences in 2011, 2013, and 2017, each introduc...
In biometrics, face recognition methods are achieving momentum with recent progress in the computer vision(CV). Face recognition is widely used in the identification of an individual's identity. Unfortunately, in recent research work has revealed this face biometrics system is unprotected to spoofing attacks using by very low price instrument such...
In biometrics, face recognition methods are achieving momentum with recent progress in the computer vision(CV). Face recognition is widely used in the identification of an individual's identity. Unfortunately, in recent research work has revealed this face biometrics system is unprotected to spoofing attacks using by very low price instrument such...
Citations
... Recently, multi-spectral LiDAR systems have been introduced to overcome this limitation in object recognition by providing additional material information based on spectroscopic imaging [27][28][29][30][31][32] . Especially the reflection spectrum in the short-wave infrared (SWIR) range (900-2500 nm wavelength) provides more comprehensive information of the material properties compared to visible-range spectroscopy or simple color imaging of the object [32][33][34][35] . SWIR range multi-spectral LiDAR systems have demonstrated enhanced identification and recognition capabilities by simultaneous acquisition of spatial and spectral information. ...
... The reflection spectrum shows clearly distinguishable differences depending on the material. In particular, the human skin is not only clearly distinguished from other objects but also exhibits a similar spectral response among numerous different test subjects, which is consistent with previous research 34 . This result shows that multi-spectral measurements in the SWIR range can provide sufficient information to classify the reflecting material. ...
The recent progress in the development of measurement systems for autonomous recognition had a substantial impact on emerging technology in numerous fields, especially robotics and automotive applications. In particular, time-of-flight (TOF) based light detection and ranging (LiDAR) systems enable to map the surrounding environmental information over long distances and with high accuracy. The combination of advanced LiDAR with an artificial intelligence platform allows enhanced object recognition and classification, which however still suffers from limitations of inaccuracy and misidentification. Recently, multi-spectral LiDAR systems have been employed to increase the object recognition performance by additionally providing material information in the short-wave infrared (SWIR) range where the reflection spectrum characteristics are typically very sensitive to material properties. However, previous multi-spectral LiDAR systems utilized band-pass filters or complex dispersive optical systems and even required multiple photodetectors, adding complexity and cost. In this work, we propose a time-division-multiplexing (TDM) based multi-spectral LiDAR system for semantic object inference by the simultaneous acquisition of spatial and spectral information. By utilizing the TDM method, we enable the simultaneous acquisition of spatial and spectral information as well as a TOF based distance map with minimized optical loss using only a single photodetector. Our LiDAR system utilizes nanosecond pulses of five different wavelengths in the SWIR range to acquire sufficient material information in addition to 3D spatial information. To demonstrate the recognition performance, we map the multi-spectral image from a human hand, a mannequin hand, a fabric gloved hand, a nitrile gloved hand, and a printed human hand onto an RGB-color encoded image, which clearly visualizes spectral differences as RGB color depending on the material while having a similar shape. Additionally, the classification performance of the multi-spectral image is demonstrated with a convolution neural network (CNN) model using the full multi-spectral data set. Our work presents a compact novel spectroscopic LiDAR system, which provides increased recognition performance and thus a great potential to improve safety and reliability in autonomous driving.
... Examples include a safety feature on a drilling machine and a guard on a table saw (a mechanism blocks the saw blade when skin is near the cutting area). Steiner et al. [11,12] transferred the concept to an imaging system by employing a dedicated camera with an InGaAs-based sensor instead of single photodiodes. The proposed system is able to combine face biometrics and PAD with the same detector. ...
Due to their user-friendliness and reliability, biometric systems have taken a central role in everyday digital identity management for all kinds of private, financial and governmental applications with increasing security requirements. A central security aspect of unsupervised biometric authentication systems is the presentation attack detection (PAD) mechanism, which defines the robustness to fake or altered biometric features. Artifacts like photos, artificial fingers, face masks and fake iris contact lenses are a general security threat for all biometric modalities. The Biometric Evaluation Center of the Institute of Safety and Security Research (ISF) at the University of Applied Sciences Bonn-Rhein-Sieg has specialized in the development of a near-infrared (NIR)-based contact-less detection technology that can distinguish between human skin and most artifact materials. This technology is highly adaptable and has already been successfully integrated into fingerprint scanners, face recognition devices and hand vein scanners. In this work, we introduce a cutting-edge, miniaturized near-infrared presentation attack detection (NIR-PAD) device. It includes an innovative signal processing chain and an integrated distance measurement feature to boost both reliability and resilience. We detail the device’s modular configuration and conceptual decisions, highlighting its suitability as a versatile platform for sensor fusion and seamless integration into future biometric systems. This paper elucidates the technological foundations and conceptual framework of the NIR-PAD reference platform, alongside an exploration of its potential applications and prospective enhancements.
... Most conventional image-based algorithms have an excellent performance in terms of accuracy when the face image is recorded under controlled conditions. However, these methods fail when presented with images captured under ✓ RGB-D Face (VAP) [69] ✓ FRGCv2 [12] ✓ BU-3DFE [15] ✓ BU-4DFE [17] ✓ 3D-TEC [87] ✓ UMB-DB [88] ✓ SuperFaces [89] ✓ ✓ CurtainFaces [70], [90] ✓ FaceWarehouse [71] ✓ Lock3DFace [43] ✓ IIIT-D RGB-D [72] ✓ UHDB11 [91] ✓ CAS(ME) 3 [92] ✓ 3DMAD [34], [35] ✓ CASIA-SURF [79] ✓ WMCA [74] ✓ HQ-WMCA [93] ✓ CASIA-SURF CeFA [94] ✓ ND-2006 [95] ✓ GavabDB [13] ✓ UoY [96] ✓ BJUT-3D [19] ✓ FRAV3D [14] ✓ Pandora [47] ✓ MPIBC [6] ✓ BIWI [97] ✓ ICT-3DHP [98] ✓ KaspaAROV [82] ✓ ✓ HRRFaceD [83] ✓ IST-EURECOM LFFD [46] ✓ FaceVerse-Detailed [99] ✓ FaceVerse-Coarse [99] ✓ Cui et al. [100] ✓ ESRC3D [49] ✓ SURREY [101] ✓ JNU [101] ✓ 3DWF (2019) [102] ✓ Intellifusion [103], [104] Li et al. [105] ✓ SeetaFace [106] ✓ MotorMark [107] ✓ Sun et al. [108] ✓ IAS-Lab [109] ✓ RGBDFaces [110] ✓ MICC (Florence2D/3D) [111] ✓ Face-Emotion [112] ✓ Florence3D-Re-Id [84] ✓ IKFDB [113] ✓ MMFD [114] ✓ RGB-D-T [115] ✓ FIDENTIS [50] ✓ FaceScape [116] ✓ CASIA HFB [21] ✓ 4DFAB [51] ✓ ✓ ND-Collection-D [10] ✓ 3DFACE-XMU [117] ✓ ZJU-3DFED [118] ✓ FSU [119] ✓ B3D(AC) [25] ✓ ✓ D3DFACS [30] ✓ Hi4D-ADSIP [31] ✓ ADSIP [22] ✓ MAVFER [56] ✓ HeadSpace [57] ✓ MeIn3D [45] ✓ Tuft [58] ✓ UHDB31 [48] ✓ Bechman [120] ✓ Eurocom [53] ✓ ✓ ✓ Msspoof [126] ✓ ✓ SWIR [127] ✓ ✓ ✓ BRSU [128] ✓ ✓ EMSPAD [129] ✓ MLFP [130] ✓ ✓ CASIA-SURF [79] ✓ CIGIT-PPM [131] ✓ ✓ PolyU-HSFD [24] ✓ CMU-HSFD [8] ✓ ND-Collection-C [9] ✓ ✓ ND-NIVL [44] ✓ ✓ CASIA HFB [21] ✓ ✓ CASIA NIR-VIS [32] ✓ ✓ LDHF-DB [36] ✓ ✓ NFRAD [29] ✓ ✓ PolyU-NIRFD [132] ✓ ✓ NVIE [26] ✓ ✓ Liu et al. [42] ✓ ✓ IRIS [133] ✓ ✓ UH [134] ✓ Carl [27] ✓ ✓ ARL-MMFD1 [135] ✓ ✓ ARL-MMFD2 [136] ✓ ✓ UL-FMTV [52] ✓ Eurocom [53] ✓ ✓ Tuft [58] ✓ ✓ ✓ Sejong-A [59] ✓ ✓ ✓ Sejong-B [59] ✓ ✓ ✓ Sober Drunk [38], [39] ✓ PUCV-DTF [54] ✓ TFW [137] ✓ ✓ SpeakingFaces [138] ✓ ✓ KTFE [41] ✓ ✓ NIST/Equinox [139] ✓ ✓ SDFD [55] ✓ CBSR-NIR [140] ✓ ✓ RWTH [141] ✓ UNCC-ThermalFace [60] ✓ IRIS-M3 [16] ✓ UWA-HSFD [142] ✓ an uncontrolled environment with high distortions resulting from changes in illumination. A nighttime situation is an example of a condition where human recognition, based exclusively on visible spectrum pictures, may be impractical. ...
... It covers genuine face images, printed VIS, and NIR images. BRSU [128] consists of 130 participants and combines spectral measurements at several points on faces and limbs with pictures taken using both an RGB camera and the presented multispectral camera system. The Extended Multispectral Presentation Attack Face Dataset, EMSPAD [129], comprises face scans of 50 subject collected by a multispectral camera for both the evaluation of presentation attack detection and the analysis of face presentation attack vulnerability. ...
Due to their ease-of-use, biometric verification methods to control access to digital devices have become ubiquitous. Many rely on supervised machine learning, a process that is notoriously data-hungry. At the same time, biometric data is sensitive from a privacy perspective, and a comprehensive review from a data set perspective is lacking. In this survey, we present a comprehensive review of multimodal face data sets (e.g., data sets containing RGB color plus other channels such as infrared or depth). This follows a trend in both industry and academia to use such additional modalities to improve the robustness and reliability of the resulting biometric verification systems. Furthermore, such data sets open the path to a plethora of additional applications, such as 3D face reconstruction (e.g., to create avatars for VR and AR environments), face detection, registration, alignment, and recognition systems, emotion detection, anti-spoofing, etc. We also provide information regarding the data acquisition setup and data attributes (ethnicities, poses, facial expressions, age, population size, etc.) as well as a thorough discussion of related applications and state-of-the-art benchmarking. Readers may thus use this survey as a tool to navigate the existing data sets both from the application and data set perspective. To existing surveys we contribute, to the best of our knowledge, the first exhaustive review of multimodalities in these data sets.
... However, sacrificing FOV is a non-optimal solution. More broadly, flash illumination or active illumination is also often used in other imaging modalities, where the environmental light level is low, including active hyper-spectral, short-wave infrared [57], middle-wave infrared [36], long-wave infrared [35], and near-ultraviolet imaging. These techniques also face the challenge of light fall off, and they may face more challenges than active, visible-light imaging in that the sensor qualities [54] and light emitters' optical efficiencies [61] are lower, or the power is constrained by safety concerns. ...
Flash illumination is widely used in imaging under low-light environments. However, illumination intensity falls off with propagation distance quadratically, which poses significant challenges for flash imaging at a long distance. We propose a new flash technique, named ``patterned flash'', for flash imaging at a long distance. Patterned flash concentrates optical power into a dot array. Compared with the conventional uniform flash where the signal is overwhelmed by the noise everywhere, patterned flash provides stronger signals at sparsely distributed points across the field of view to ensure the signals at those points stand out from the sensor noise. This enables post-processing to resolve important objects and details. Additionally, the patterned flash projects texture onto the scene, which can be treated as a structured light system for depth perception. Given the novel system, we develop a joint image reconstruction and depth estimation algorithm with a convolutional neural network. We build a hardware prototype and test the proposed flash technique on various scenes. The experimental results demonstrate that our patterned flash has significantly better performance at long distances in low-light environments.
... Hyperspectral imaging acquires images with hundreds of continuous wavebands usually by the application of a spectrographs and a sensitive area detector (Qin et al., 2013;Gonzalez et al., 2018). In the case of multispectral imaging a sensitive area detector is usually paired with a or series of specific waveband filters or a waveband tunable light source (Steiner et al., 2016). The result of both systems is called a data cube, which is composed of a M × N × W matrix, where M and N represent the ☆ This work was supported by Cooperative Research Centres Projects program (CRCPSIX000081), Australia and Surenut Pty Ltd. ...
The design of a multispectral camera system to replace hyperspectral cameras used in chemometrics laboratory for food research is a challenging task in the field of machine vision. In this paper, the process of designing and prototyping a low cost multispectral camera to measure aflatoxin B1 in almonds is described. A multispectral system with a Goldeye G-008 TEC1 SWIR is developed to replace a hyperspectral system with a Specim FX17 line scanner. Using R-squared and RMSE values of multiple linear regression, the performance of a hyperspectral camera system with a multispectral camera system were compared for measuring aflatoxin B1.
The results show that a high speed multispectral system can replace an equivalent hyperspectral imaging system with much lower cost with an acceptable processing performance of 20 datacubes per seconds.
... Median filter can be used to eradicate noise from the image if the camera is high definition, however occasionally it still has noise. Skin classification was implemented; this changes the whole pixel to black excluding the pixel which are near the skin [12]. Proposed International Automatic Attendance Management System which uses Facial Recognition. ...
Attendance is an essential aspect of learning process in every tertiary institution. Attendance taking in every class is a day to day activity in a tertiary institutions and organisations. The traditional ways of taken the student attendance by signing of papers or calling of students name in the class is also time consuming and unconfident. The contemporary academic procedure of repeating or calling names of student in a class attendance compete a substantial role in eminence of teaches and performance evaluation of the students. The administration of the attendance may also lead to enormous problem if administer manually. This paper intends to design attendance monitoring system using artificial intelligent. To solve the problem of attendance in class, camera will be used for capturing faces of student individually; recognize each student and update the database accordingly. Face geometry algorithm, features invariant and machine learning based methods will be applied to solve the problem. Extraction and pre-processing of face region is conducted for advanced processing. Resizing and extraction of face image involves histogram equalization and pre-processing. The image contrast is improved and clearer, since the image intensity is stretches.
... Recently, there have been several CNN based methods that leverage multi-channel information for PAD. In addition to legacy RGB data, several works have suggested the use of depth, thermal [27], [35], [36], [37], [38], near-infrared (NIR), shortwave infrared (SWIR) [30], [39], laser speckle imaging, light field imaging [40], [41] and so on. However, the effectiveness of these channels on a wide variety of attacks has not been studied comprehensively. ...
The vulnerability against presentation attacks is a crucial problem undermining the wide-deployment of face recognition systems. Though presentation attack detection (PAD) systems try to address this problem, the lack of generalization and robustness continues to be a major concern. Several works have shown that using multi-channel PAD systems could alleviate this vulnerability and result in more robust systems. However, there is a wide selection of channels available for a PAD system such as RGB, Near Infrared, Shortwave Infrared, Depth, and Thermal sensors. Having a lot of sensors increases the cost of the system, and therefore an understanding of the performance of different sensors against a wide variety of attacks is necessary while selecting the modalities. In this work, we perform a comprehensive study to understand the effectiveness of various imaging modalities for PAD. The studies are performed on a multi-channel PAD dataset, collected with 14 different sensing modalities considering a wide range of 2D, 3D, and partial attacks. We used the multi-channel convolutional network-based architecture, which uses pixel-wise binary supervision. The model has been evaluated with different combinations of channels, and different image qualities on a variety of challenging known and unknown attack protocols. The results reveal interesting trends and can act as pointers for sensor selection for safety-critical presentation attack detection systems. The source codes and protocols to reproduce the results are made available publicly making it possible to extend this work to other architectures.
... There are a total of 681 images for each modality, with at least one frontal face image, and from five to nine frontal disguised images per subject. The BRSU Spoof Database [24,25] is a multispectral DB, with images captured in visible and infrared modalities at frequencies of 935, 1060, 1200, and 1550 nm. There are several variations in the DB that render it challenging, such as expression, makeup, 3D masks, fake beards, glasses, fake noses, and presentation attacks. ...
Applications for facial recognition have eased the process of personal identification. However, there are increasing concerns about the performance of these systems against the challenges of presentation attacks, spoofing, and disguises. One of the reasons for the lack of a robustness of facial recognition algorithms in these challenges is the limited amount of suitable training data. This lack of training data can be addressed by creating a database with the subjects having several disguises, but this is an expensive process. Another approach is to use generative adversarial networks to synthesize facial images with the required disguise add-ons. In this paper, we present a synthetic disguised face database for the training and evaluation of robust facial recognition algorithms. Furthermore, we present a methodology for generating synthetic facial images for the desired disguise add-ons. Cycle-consistency loss is used to generate facial images with disguises, e.g., fake beards, makeup, and glasses, from normal face images. Additionally, an automated filtering scheme is presented for automated data filtering from the synthesized faces. Finally, facial recognition experiments are performed on the proposed synthetic data to show the efficacy of the proposed methodology and the presented database. Training on the proposed database achieves an improvement in the rank-1 recognition rate (68.3%), over a model trained on the original nondisguised face images.
... The BRSU Spoof Database [14], [15] contains images captured in the visible and infrared modalities at four different spectral frequencies (935 nm, 1060 nm, 1300 nm, and 1550 nm). The database contains variations in expression, makeup, 3D masks, fake beard, glasses, fake nose and presentation attack. ...
Commercial application of facial recognition demands robustness to a variety of challenges such as illumination, occlusion, spoofing, disguise, etc. Disguised face recognition is one of the emerging issues for access control systems, such as security checkpoints at the borders. However, the lack of availability of face databases with a variety of disguise addons limits the development of academic research in the area. In this paper, we present a multimodal disguised face dataset to facilitate the disguised face recognition research. The presented database contains 8 facial add-ons and 7 additional combinations of these add-ons to create a variety of disguised face images. Each facial image is captured in visible, visible plus infrared, infrared, and thermal spectra. Specifically, the database contains 100 subjects divided into subset-A (30 subjects, 1 image per modality) and subset-B (70 subjects, 5 plus images per modality). We also present baseline face detection results performed on the proposed database to provide reference results and compare the performance in different modalities. Qualitative and quantitative analysis is performed to evaluate the challenging nature of disguise addons. The dataset will be publicly available with the acceptance of the research article. The database is available at: https://github.com/usmancheema89/SejongFaceDatabase.
... The BRSU Spoof Database [14], [15] contains images captured in the visible and infrared modalities at four different spectral frequencies (935 nm, 1060 nm, 1300 nm, and 1550 nm). The database contains variations in expression, makeup, 3D masks, fake beard, glasses, fake nose and presentation attack. ...
Commercial application of facial recognition demands robustness to a variety of challenges such as illumination, occlusion, spoofing, disguise, etc. Disguised face recognition is one of the emerging issues for access control systems, such as security checkpoints at the borders. However, the lack of availability of face databases with a variety of disguise add-ons limits the development of academic research in the area. In this paper, we present a multi-modal disguised face dataset to facilitate the disguised face recognition research. The presented database contains 8 facial add-ons and 7 additional combinations of these add-ons to create a variety of disguised face images. Each facial image is captured in visible, visible plus infrared, infrared, and thermal spectra. Specifically, the database contains 100 subjects divided into Subset-A (30 subjects, 1 image per modality) and Subset-B (70 subjects, 5 plus images per modality). We also present baseline face detection results performed on the proposed database to provide reference results and compare the performance in different modalities. Qualitative and quantitative analysis is performed to evaluate the challenging nature of disguise add-ons. The dataset will be publicly available with the acceptance of the research article.
